Last week, various corners of the mediamonde asserted that the Dodd-Frank Act's repeal of Rule 436(g), under the Securities Act of 1933, caught Nationally Recognized Statistical Rating Organizations (NRSROs) like Standard and Poor's, Moody's Investors Service and Fitch Ratings by surprise.
In brief, the repeal exposes NRSROs to the same liability, under Sections 7 and 11 of the 1933 Act, as other "experts" (such as lawyers, accountants, engineers or other professionals) whose statements or reports are used in connection with a security registration statement (e.g. prospectus).
The top three NRSROs retaliated by, as the Wall Street Journal puts it, asking their clients, "Please don't use our credit ratings." That is, they will continue to rate issues, but they will no longer consent for their use in SEC filings as other experts must. (Providing written consent, which is duly filed with the SEC, is a formality prescribed under the 1933 Act.)
Unfortunately, new public ABS issues require a rating. The alternative is not to issue, or to issue in the smaller, less liquid private placement market under Rule 144A. A few non-mortgage public ABS deals already in the works were withdrawn by their sponsors and the usual suspects proclaimed Congress had achieved the unintended consequence of smothering the ABS market and stifling consumer credit.
Typical of the "surprise camp", WSJ reporter Anusha Shrivastava, wrote on July 21 that the change startled NRSROs because the provision was not included in the original Senate version of the bill. It may make brighter prose to call it a surprise, but the NRSROs have been grappling with the possible repeal of Rule 436(g) for many many months now. You can be sure they had their Washington consultants and lobbyists on the job throughout the drafting, debate, voting and reconciling of the legislation.
It's more likely that mediacrats were surprised. And no wonder, given the size of the bills brought to conference in June, it was tough enough to keep track of the big splash items, like the Financial Stability Oversight Council, the Volker rule or OTC derivatives overhaul.
The fact is, the original Senate financial reform bill contained other provisions aimed at the NRSROs, but not the repeal of the 436(g) exemption. More to the point, the provision was in the House bill, which put it on the table for the conference committee's report (that is, the final version of the law revoted and sent to the President). The only change to the House provision was its effective date. The original House provision went into effect after six months, but no date was indicated in the Dodd-Frank Act. That automatically made it effective one day after the President signed it into law, July 22.
The Conference Committee completed its work in the wee hours of June 25, 2010, plenty of time for all interested parties to spot the provision and take their case to the SEC and into the media. The surprise is that the "story broke" so late - after the Senate's approved it on July 15.
Stay of Execution
Another good sign the repeal was not a surprise was the SEC's timely response. The day the provision became effective the SEC announced a six month reprieve aimed directly at the ABS market. In a statement by Meredith Cross, Director of the Division of Corporation Finance, the Commission acknowledged that it had been reviewing the interaction between its Regulation AB requirements for ABS and the requirement for rating agencies to consent to use of their ratings in filings.
Accordingly, it was issuing a "no action" letter allowing ABS issuers to omit credit ratings from registration statement for a period of six months. In its no-action letter to Ford Motor Credit Company, July 22, 2010, SEC explained that Reg AB requires disclosure of whether the issuance or sale of any class of the offered ABS is conditioned on the assignment of a rating by one or more raters. By conditioned, SEC is referring to the fact that the rating is provisional until the actual collateral is delivered to the trust at closing, usually days or weeks after the marketing period. If so conditioned, then Reg AB requires disclosure of the minimum credit rating that must be assigned and the identity of the rater(s). Provisions to monitor ratings during the term of the ABS must also be disclosed.
The Commission also clarified a point overlooked in most press and blog accounts of the ratings kerfuffle - that 436(g) covers all inclusions of NSRSO ratings in SEC filings, not just those of ABS issuers. However, it explained that the rules for corporate debt issuances differ from those for ABS issues. To address those differences it published a handful of new compliance and disclosure interpretations. Corporate issuers who include ratings in their registration statements would not have to file a consent by the rater until the next time they are amended (for instance, in the next annual report 10-K). Furthermore, a consent would not be required if the credit rating in the filing is related to a change in credit rating, the registrant's liquidity or cost of funds. Nor would it be needed with a free writing prospectus, term sheet or press release otherwise compliant with SEC rules.
Otherwise, corporate issuers' registration statements or amendments effective after July 22, 2010 will require the raters consent. The difference is that, although corporate issuers commonly disclose ratings, they are not required to do so.
Long Time Coming
Stripping ratings purveyors of their Rule 436(g) protections has been under discussion for some time. In fact, the SEC seriously considered doing so itself last year, publishing a "concept release" on the possible rescission of 436(g) last October.
I found the background discussion in the concept release helpful in highlighting the issues. As the Securities and Exchange Commission explains, Section 11 of the Securities Act was intended to subject to a rigorous standard of liability those persons with a direct role in a registered offering:
... to assure that disclosure regarding securities is accurate. It was also designed to give investors additional protection not available under common law due to the barriers to recovery presented by the common law fraud requirements of scienter, reliance and causation. Liability under Section 11 extends to the issuer, officers and directors who sign the registration statement, underwriters, and persons who prepare or certify any part of the registration statement or who are named as having prepared or certified a report or valuation for use in connection with the registration statement.
An expert may be held liable for an untrue statement or omission of material fact "unless he can establish that he had, after reasonable investigation, reasonable grounds to believe and did believe" that the statements were true and material facts needed to make the statements not misleading had not been omitted.
(This is not all that different from the standard under which as a sell side analyst, I published my analysis, commentary and recommendations. Violating that standard risks firing and being barred from the securities industry. I am not distressed to see it imposed on the raters.)
Rule 436(g) was adopted in 1982, after the SEC reversed a long standing policy of discouraging the disclosure of credit ratings. The rationale for for discouraging disclosure of ratings is difficult to ferret out of the SEC site. The 1977 concept release 33-5882, proposing the change, appears not to be available electronically. Subsequent discussion, in SEC 33-6336 (I found on lawyerlinks.com, not SEC.gov) for instance, suggest that a common objection to ratings is that, by compressing too much information into a single grade, they could confuse investors.
The shift was simply to allow voluntary disclosure. In proposing the change in policy, the SEC asked comments on whether raters should be treated like other experts under Section 11. The 2009 release indicates that comments generally opposed subjecting NRSROs to liability because, among other things, it would interfere with the substance and timing of the registration process, it would change the way credit ratings were issued, and it would result in higher costs and increased uncertainty over the scope of liability.
In particular, "The NRSROs in existence in 1977 indicated that they would not provide consents to be named in the registration statement." Subsequent comments revealed concerns that the raters independence would be affected if they were "participants" in the offering. And, get this - the quality of ratings would be diminished because raters "would rely only on objective, quantifiable information." Hunches, gut guesses, fingers to the wind and wishful thinking would not be admissible.
Please note, these arguments were advanced long before the rise of private MBS and ABS markets. In fact, the discussion took place even before the junk bond market took off. Even then, however, the issues were the same as they are now. The commenters who favored subjecting NRSROs to liability under Section 11 argued that doing so would incline them to take more care in determining ratings.
Whatever. Rule 436(g) was adopted.
Will the Real Expert Please Stand Up
Fast forward to last year's SEC proposal to rescind 436(d). One of the arguments the Commission made for doing so was that NSRSOs "represent themselves to registrants and investors as experts at analyzing credit and risk." Although the raters describe their credit ratings as "opinions" protected under the First Amendment, many other professionals provide legal opinions, valuations opinions, fairness opinions and audit reports that are incorporated in filings and on which investors rely. Those professionals are treated as "experts" and are subject to the consent requirements under the Securities Act.
Furthermore, many types of professionals can rate the claims paying ability of a bond issuer (in the case of an ABS, the issuer is the legal structure - e.g. a limited purpose corporation, trust or combination of such entities - created to issue bonds to buy assets used to service the bonds). However, only NRSROs enjoyed the 436(g) exemption - the others, regardless of their track records, market shares, marketing clout and so forth, were held accountable as experts.
Consider this scenario. As a consultant with long experience of various sorts in structured products, I hang out my "rater" shingle and a lender retains my services on an asset-backed issue. I rate the bonds triple-A, with a very high probability that all interest and principle payments will be made as described in the prospectus. I am an expert and must consent to the use of my opinion. And it better be the best, most fact-based and defensible opinion I can devise, because I am legally liable.
Investors might respect my research over the years, there could even be those who made investment decisions under its influence, but there is not one, currently active or retired from the market, who would invest in those bonds based on an understanding that I have staked my reputation and business and affordable liability insurance premiums on that rating. On an understanding that my conscience is as invested in that rating as the lawyer's is in her true sale opinion. A huge NRSRO, in the rating business for decades, with published criteria, the budget for historical asset performance data, legions of analysts and a credit committee to review their work rates it as well, but does not consent to be treated as an expert.
Step Up and Be Experts
My hope is that the raters use the next six months to adjust to their newly legislated expertise. Big picture, in a world where liabilities trump responsibilities, it seems to me that if obstetricians close their practices because they can't afford malpractice insurance, it's only fair to make rating agencies liable as the experts they want to be paid for being. And how handsomely they get paid has already been read into the public record.
Now that rescinding 436(g) is a done deal, it should be possible for the ratings firms' legal counsel to form a less reactive picture of the real, expanded potential for law suits. After all, the language of Section 11 (summarized above) has given comfort to the accountants, attorneys and other experts who do give their consent.
I'm not the only structured product analyst to read the 1933 Act as comforting, either. Alan Todd, Matthew Jozoff and Amy Sze at JP Morgan Securities write in their weekly that the wording of Section 11 suggests that "at the time a deal is issued, rating agency personnel, which base their ratings on both empirical as well as forward looking scenarios, need only prove that they believed their results to be accurate, complete and not misleading."
NOTE: Linda Lowell writes a regular column, called Kitchen Sink, for HousingWire magazine.
Editor’s note: Linda Lowell is a 20-year-plus veteran of MBS and ABS research at a handful of Wall Street firms. She is currently principal of OffStreet Research LLC.